Japanese Transitive and Intransitive Verbs: e-Learning System with Speech Recognition and Video
Hiroaki Okumoto, Waseda University, Japan
EdMedia + Innovate Learning, in Lugano, Switzerland ISBN 978-1-880094-53-2 Publisher: Association for the Advancement of Computing in Education (AACE), Waynesville, NC
Learning a foreign language requires diverse knowledge and skills. So we developed a Web-based learning system. This system is equipped with a speaker-independent speech recognition applet, so learners can use voice input without installing any additional software. The speech recognition system uses formant frequency, logarithmic power and spectral power ratio as phonetic feature, and analyzes them by both DTW and HMM. The system offers such exercises as voicing the answer and filling in conversation in video footages. We had international students use this system and collected their evaluation using a questionnaire. 65% of them replied that this system was helpful. Nearly 70% of learners were interested in a learning tool with speech recognition and more than 90% thought it important to pay attention to pronunciation. From these results, we reached the conclusion that voice-based learning tool is beneficial to improve both learners' motivation and learning effectiveness.
Okumoto, H. (2004). Japanese Transitive and Intransitive Verbs: e-Learning System with Speech Recognition and Video. In L. Cantoni & C. McLoughlin (Eds.), Proceedings of ED-MEDIA 2004--World Conference on Educational Multimedia, Hypermedia & Telecommunications (pp. 1902-1907). Lugano, Switzerland: Association for the Advancement of Computing in Education (AACE).
© 2004 Association for the Advancement of Computing in Education (AACE)